With the rapid development of remote sensing technology, the application of optical remote sensing images in productivity estimation is becoming increasingly widespread. However, due to the limited temporal and spatial resolution of remote sensing images, accurately estimating productivity still faces challenges. This study improves the spatiotemporal resolution of optical remote sensing images through spatiotemporal fusion algorithms to achieve more accurate productivity estimation. We use advanced spatiotemporal fusion technology to combine high temporal resolution but low spatial resolution images with high spatial resolution but low temporal resolution images to generate composite images with high spatiotemporal resolution. These composite images can more accurately capture the dynamic changes in surface productivity. Even at the longest time scale of 3 months, the spectral fidelity remains at a high level of 0.87, indicating that the model still has good spectral fidelity performance when processing remote sensing images with long time intervals. This study not only improves the accuracy of productivity estimation, but also provides new data support for resource management and decision-making in industries such as agriculture and forestry, which has important theoretical and practical significance. In addition, our method can also provide reference for other fields and promote the widespread application and development of remote sensing technology.